https://github.com/rose-stl-lab/limo-plus
Follow-up study of LIMO by Vineet Thumuluri
https://github.com/rose-stl-lab/limo-plus
Last synced: over 1 year ago
JSON representation
Follow-up study of LIMO by Vineet Thumuluri
- Host: GitHub
- URL: https://github.com/rose-stl-lab/limo-plus
- Owner: Rose-STL-Lab
- Created: 2023-07-01T22:08:59.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2024-07-23T15:43:44.000Z (almost 2 years ago)
- Last Synced: 2025-01-04T12:16:44.796Z (over 1 year ago)
- Language: Python
- Homepage:
- Size: 217 MB
- Stars: 0
- Watchers: 0
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Variants of the Latent Inceptionism on Molecules (LIMO) framework to improve the properties of generated molecules.
# Technical Report
[Technical report: Improving the properties of molecules generated by LIMO](https://arxiv.org/abs/2407.14968)
# Experiments with LIMO
## Installation
Please ensure that [RDKit](https://www.rdkit.org/docs/Install.html) and [Open Babel](https://openbabel.org/wiki/Category:Installation) are installed. The following Python packages are also required (these can also be installed with `pip install -r requirements.txt`):
```
torch
pytorch-lightning==1.9.0
selfies
scipy
tqdm
```
Code was tested with Python 3.9, but will likely work on any version of Python 3.
## Tokenizer config
Use `src/tokenizers.py` script to generate data using different tokenizers.
## Training + generation
Use the `run_limo.py` script with different arguments to run different stages (training, fine-tuning, generation), different tokenizer configurations, different optimization strategies etc.